The use of Relational Database Management Systems (RDBMS), a type of microcomputer application software, to analyze open-ended survey questions is discussed. Using open-ended questions allows researchers to ask respondents to express themselves freely about their attitudes and beliefs. This approach also can elicit a precise answer even though the list of possible answers is too large to list in the instrument or too long for most respondents to read (such as student major or home town). The characteristics of an RDBMS that allow for analysis of open-ended questions are: flexibility to create fields after the database has been designed, ability to join databases, and ability to sort on any field in the database. RDBMS can handle unstructured data and can use the relational operators "join" and "project" when using CONDOR, or the relational operator "copy" when using dBASE. The RDBMS can be used to clarify who is responding to open-ended questions in surveys, thus making the comments more useful, even in cases of underenumeration. The underenumeration problem can also be approached through effective design techniques. Another feature of RDBMS is that it allows the creation of a data entry screen. In addition to identifying technical considerations related to the use of RDBMS, an example of the automated Q-sort is provided. (SW)